Compartmentalized Linker Array: A Scalable and Transferrable Microarray Format for Multiplexed Immunoassays.

Microarrays have been widely used for multiplexed bioassays. Fabrication of a conventional microarray typically requires a complex microarray spotter, using which nanoliter bioreagent (e.g., antibody and cells) droplets are delivered onto a glass slide. However, arraying a delicate bioreagent in nanoliter volumes could cause the loss of bioactivity and needs a complex microarray spotter. Further, mixing of different bioreagents in a multiplexed assay leads to cross-reactions, producing false positive signals that impair assay reproducibility and scalability. In this work, we propose a new microarray format, named "compartmentalized linker array (CLA)", that consists of pre-prepared storable microarrays of chemical linkers in microliter compartments. CLA can be used for binding and patterning bioreagents into microarrays by simply pipetting and incubating bioreagent solutions in compartments. Using commonly used aminosilane linker-based antibody microarray, we developed CLA and demonstrated its application for a multiplexed sandwich immunoassay measuring three cancer-related proteins. A "two-phase" blocking system was established for de-activating background regions on glass where no linker molecules are present. Storage conditions of the CLA chip were explored and demonstrated for long-term storage. In a multiplexed immunoassay, low pg/mL sensitivity was achieved for all the three proteins, comparable to those of conventional assays. Moreover, CLA can be potentially used for other applications beyond protein assays, making microarray technology transferrable and widely available for the biological and biomedical research community.

In-silico selection of cancer blood plasma proteins by integrating genomic and proteomic databases.

Blood protein markers have been studied for the clinical management of cancer. Due to the large number of the proteins existing in blood, it is often necessary to pre-select potential protein markers before experimental studies. However, to date there is a lack of automated method for in-silico selection of cancer blood proteins that integrates the information from both genetic and proteomic studies in a cancer-specific manner. In this work, we synthesized both genomic and proteomic information from several open access databases and established a bioinformatic pipeline for in-silico selection of blood plasma proteins overexpressed in specific type of cancer. We demonstrated the workflow of this pipeline with an example of breast cancer, while the methodology was applicable for other cancer types. With this pipeline we obtained 10 candidate biomarkers for breast cancer. The proposed pipeline provides a useful and convenient tool for in-silico selection of candidate blood protein biomarkers for a variety of cancer research.